Adequate feeding and the usefulness of the respiratory quotient in critically ill children

We determined incidences of underfeeding and overfeeding in children who were admitted to a multidisciplinary tertiary pediatric intensive care and evaluated the usefulness of the respiratory quotient (RQ) obtained from indirect calorimetry to assess feeding adequacy. Children 18 y and younger who fulfilled the criteria for indirect calorimetry entered our prospective, observational study and were studied until day 14. Actual energy intake was recorded, compared with required energy intake (measured energy expenditure plus 10%), and classified as underfeeding ( <90% of required), adequate feeding (90% to 110% of required), or overfeeding (>110% of required). We also evaluated the adequacy of a measured RQ lower than 0.85 to identify underfeeding, and an RQ higher than 1.0 to identify overfeeding. Ninety-eight children underwent 195 calorimetric measurements. Underfeeding, adequate feeding, and overfeeding occurred on 21%, 10%, and 69% of days, respectively. An RQ lower than 0.85 to identify underfeeding showed low sensitivity (63%), high specificity (89%), and high negative predictive value (90%). An RQ higher than 1.0 to indicate overfeeding showed poor sensitivity (21%), but a high specificity (97%) and a high positive predictive value (93%). Food composition, notably high-carbohydrate intake, was responsible for an RQ exceeding 1.0 in the overfed group. Children admitted to the intensive care unit receive adequate feeding on only 10% of measurement days during the first 2 wk of admission. The usefulness of RQ to monitor feeding adequacy is limited to identifying (carbohydrate) overfeeding and excluding underfeeding